BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//scienceofintelligence.de - ECPv6.16.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://www.scienceofintelligence.de
X-WR-CALDESC:Events for scienceofintelligence.de
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Europe/Berlin
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20190331T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20191027T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20200329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20201025T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20210328T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20211031T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20220327T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20221030T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20230326T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20231029T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20220310T160000
DTEND;TZID=Europe/Berlin:20220324T170000
DTSTAMP:20260619T135227
CREATED:20220221T112423Z
LAST-MODIFIED:20250604T092603Z
UID:11712-1646928000-1648141200@www.scienceofintelligence.de
SUMMARY:Mark Nawrot (North Dakota University)\, “Pursuit Eye Movements in the Perception of Depth From Motion Parallax”
DESCRIPTION:Abstract: \nThe brain performs critical calculations on visual information as we swiftly\, yet effortlessly\, navigate around objects and obstacles in our cluttered environment. Perhaps one of the most important calculations is for the perception of depth using the apparent relative motion of objects in the environment created by our own translation known as motion parallax. This presentation will illustrate how the visual system relies on the combination of retinal image motion with a pursuit eye movement signal to quickly determine the relative depths of objects using motion parallax. Relative depth is accurately modelled with a simple formula known as the motion-pursuit ratio. \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-afternoon-talk-with-mark-nawrot-north-dakota-university/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2022/02/adrien-converse-kCrrUx7US04-unsp.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20220303T160000
DTEND;TZID=Europe/Berlin:20220310T170000
DTSTAMP:20260619T135227
CREATED:20220221T112015Z
LAST-MODIFIED:20240813T100419Z
UID:11709-1646323200-1646931600@www.scienceofintelligence.de
SUMMARY:Chaz Firestone (Johns Hopkins University)\, "Seeing 'How'"
DESCRIPTION:Abstract: \nWhat is perception? The most intuitive and influential answer to this question has long been the one given by David Marr: To see the world is “to know what is where by looking” – to transform light into representations of objects and their features\, located somewhere ins pace. But is this all that perception delivers? Consider a figure composed by pieces of a puzzle; certainly you see some colored shapes\, as well as where they are located. Yet\, beyond this\, you may also see how they relate to one another: A\, say\, green piece can fit into the others\, and even create a new object with a shape of its own. \nIn this talk\, I present evidence that perception extracts relations between objects in much the same way as it processes the objects themselves\, and that these relations are abstract\, structured\, and surprisingly sophisticated. We’ll explore (and experience) the perception of several sophisticated relations between objects\, including combining\, supporting\, containing\, covering\, and fastening – as well as relational “illusions” in which objects appear to interact with mysteriously invisible entities. Together\, this work suggests that we see not only “what” and “where”\, but also “how”. \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-afternoon-talk-with-chaz-firestone/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2022/02/alina-grubnyak-tEVGmMaPFXk-unsplash.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20220217T100000
DTEND;TZID=Europe/Berlin:20220217T110000
DTSTAMP:20260619T135227
CREATED:20220117T152745Z
LAST-MODIFIED:20240813T100442Z
UID:11627-1645092000-1645095600@www.scienceofintelligence.de
SUMMARY:Yuejiang Liu (EPFL University)\, "Learning Beyond the IID Setting with Robust and Adaptive Representations"
DESCRIPTION:Abstract \nMachine learning models have achieved stunning successes in the IID setting. Yet\, beyond this setting\, existing models still suffer from two grand challenges: brittle under covariate shift and inefficient for knowledge transfer. In this talk\, I will introduce three approaches to tackle these challenges\, namely self-supervised learning\, causal representation learning\, and test-time training. More specifically\, I will share our recent findings on (i) incorporating prior knowledge of negative examples into representation learning\, (ii) promoting causal invariance and structure by making use of data from multiple domains\, (iii) exploiting extra information besides model parameters for effective test-time adaptation. I will show how these techniques enable deep neural networks to more robustly generalize and efficiently adapt to new environments in the motion or vision context. I will finally discuss the implications of these results on the design\, training\, and deployment of deep models for domain generalization and adaptation. Comments and feedback are more than welcome. \n  \nPaper Links \nSocial NCE: Contrastive Learning of Socially-Aware Motion Representations\, ICCV’21 \nTTT++: When Does Self-Supervised Test-Time Training Fail or Thrive? NeurIPS’21 \nCollaborative Sampling in Generative Adversarial Networks\, AAAI’20 \nTowards Robust and Adaptive Motion Forecasting: A Causal Representation Perspective\, Preprint’21 (under review) \n  \nBio \nYuejiang Liu is a PhD student at EPFL\, advised by Alexandre Alahi. His research interests center around representation learning and its applications to autonomous agents. He is particularly excited about unsupervised learning for robust generalization and efficient adaptation. \n  \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-with-yuejiang-liu-epfl-university-learning-beyond-the-iid-setting-with-robust-and-adaptive-representations/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2022/01/photo_yuejiang-e1642433180790.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20220210T100000
DTEND;TZID=Europe/Berlin:20220210T110000
DTSTAMP:20260619T135227
CREATED:20220131T105742Z
LAST-MODIFIED:20250604T092633Z
UID:11655-1644487200-1644490800@www.scienceofintelligence.de
SUMMARY:Mathilde Caron\, “Self-Supervised Learning: How To Learn From Images Without Human Annotations”
DESCRIPTION:Abstract:\nSelf-supervised learning (SSL) consists in training neural network systems without using any human annotations. Typically\, neural networks require large amounts of annotated data\, which have limited their applications in fields where accessing these annotations is expensive or difficult. Moreover\, manual annotations are biased towards a specific task and towards the annotator’s own biases\, which can result in noisy and unreliable signals. Training systems without annotations could lead to better\, more generic and robust representations. In this talk\, I will present different contributions to the fast-growing field of SSL conducted during my PhD. I will finish by discussing open questions and challenges for the future of SSL. \n  \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-with-mathilde-caron/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2022/01/carol.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20220127T100000
DTEND;TZID=Europe/Berlin:20220127T110000
DTSTAMP:20260619T135227
CREATED:20211221T062119Z
LAST-MODIFIED:20240813T100506Z
UID:11451-1643277600-1643281200@www.scienceofintelligence.de
SUMMARY:Dimitri Coelho Mollo (Science of Intelligence)\, "The Concept of Intelligence - A progress report"
DESCRIPTION:In this presentation\, I will report on the results of my work so far on the concept of intelligence\, summarising some of the main points and proposals made\, and opening the floor for open discussion about the topic. \n  \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-with-dimitri-coelho-mollo-scioi-the-concept-of-intelligence-a-progress-report/
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2020/03/Dimitri1-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20220106T100000
DTEND;TZID=Europe/Berlin:20220106T110000
DTSTAMP:20260619T135227
CREATED:20211222T105550Z
LAST-MODIFIED:20250604T092730Z
UID:11457-1641463200-1641466800@www.scienceofintelligence.de
SUMMARY:Ruben Arslan (MPI Berlin): “Bad Science vs. Open Science. The Replication Crisis and Possible Ways Out.”
DESCRIPTION:Estimates from large-scale replication projects in psychology suggest that the majority of studies from top journals do not replicate. Using commonly accepted research methods\, several academic fields amassed prolific\, seemingly coherent literatures on phenomena that do not exist\, such as extrasensory perception and depression candidate genes. Throughout the biomedical and life sciences\, data detectives keep finding highly cited papers that are riddled with errors invalidating their conclusions. Our textbooks are full of findings that do not replicate or are otherwise in serious doubt.\nAcademia as a system has issues\, but can we use the scientific method to understand and remedy them? A vibrant reform movement is seeking to do so\, but it is hard to keep track of all the suggestions to do better and tell fads from truly beneficial reforms. I outline concrete plans and paths that could lead to lasting improvements\, such as PCI Registered Reports\, the Peer Reviewer’s Openness Initiative\, post publication peer review\, and guideline and incentive setting at the journal\, hiring and funding level.\n \n  \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-ruben-arslan-mpi-berlin-personal-and-social-information-search-and-integration-for-intelligent-decisions-on-climate-action/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/png:https://www.scienceofintelligence.de/wp-content/uploads/2021/12/Screenshot-2021-12-21-at-07.12.49.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20211223T100000
DTEND;TZID=Europe/Berlin:20211223T110000
DTSTAMP:20260619T135227
CREATED:20211125T115919Z
LAST-MODIFIED:20240813T095249Z
UID:11168-1640253600-1640257200@www.scienceofintelligence.de
SUMMARY:Elke Weber (Princeton University)\, "Personal and Social Information Search and Integration for Intelligent Decisions on Climate Action"
DESCRIPTION:Abstract:\nSome of my past and current research looks at “decisions from  experience\,” i.e.\, decisions based on the personally experienced outcomes of past choices\, along the lines of reinforcement learning models and how such learning and updating is related to and differs from the way in which people and other intelligent agents use other sources of information\, e.g.\, vicarious feedback (anecdotal/social and/or in the form of statistical distributions of outcomes) or science- or model-based outcome predictions.  What happens when these different sources of forecasts of the consequences of choices disagree with each other? How do such conflicts get resolved?  How do these different ways of learning and updating over time lie at the basis of the formation and/or modification of social norms?  And how can answers to this complex of questions be put to use to motivate greater action on climate change? \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-elke-weber-princeton-university-personal-and-social-information-search-and-integration-for-intelligent-decisions-on-climate-action/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2021/11/biophoto.elke_.closeup.2020.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20211216T160000
DTEND;TZID=Europe/Berlin:20211216T173000
DTSTAMP:20260619T135227
CREATED:20211118T084943Z
LAST-MODIFIED:20250604T092741Z
UID:11010-1639670400-1639675800@www.scienceofintelligence.de
SUMMARY:Lars Chittka (Queen Mary\, University of London)\, “The Mind of a Bee”
DESCRIPTION:Abstract: Bees have a diverse instinctual repertoire that exceeds in complexity that of most vertebrates. This repertoire allows the social organisation of such feats as the construction of precisely hexagonal honeycombs\, an exact climate control system inside their home\, the provision of the hive with commodities that must be harvested over a large territory (nectar\, pollen\, resin\, and water)\, as well as a symbolic communication system that allows them to inform hive members about the location of these commodities. However\, the richness of bees’ instincts has traditionally been contrasted with the notion that bees’ small brains allow little behavioural flexibility and learning behaviour. This view has been entirely overturned in recent years\, when it was discovered that bees display abilities such as counting\, attention\, simple tool use\, learning by observation and metacognition (knowing their own knowledge). Thus\, some scholars now discuss the possibility of consciousness-like phenomena in the bees. These observations raise the obvious question of how such capacities may be implemented at a neuronal level in the miniature brains of insects. We need to understand the neural circuits\, not just the size of brain regions\, which underlie these feats. Neural network analyses show that cognitive features found in insects\, such as numerosity\, attention and categorisation-like processes\, may require only very limited neuron numbers. Using computational models of the bees’ visual and olfactory systems\, we explore whether seemingly advanced cognitive capacities might ‘pop out’ of the properties of relatively basic neural processes in the insect brain’s visual processing area\, and their connection with the mushroom bodies\, higher order learning centres in the brains of insects. \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/distinguished-speaker-series-lars-chittka-queen-mary-university-of-london-the-mind-of-a-bee/
LOCATION:TU Berlin
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/png:https://www.scienceofintelligence.de/wp-content/uploads/2021/11/Screenshot-2021-11-18-at-09.43.41.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20211216T100000
DTEND;TZID=Europe/Berlin:20211216T110000
DTSTAMP:20260619T135227
CREATED:20211125T115451Z
LAST-MODIFIED:20250604T092750Z
UID:11159-1639648800-1639652400@www.scienceofintelligence.de
SUMMARY:Romain Couillet (University Grenoble-Alps\, France)\, “Random Matrices Could Steer the Dangerous Path Taken by AI but Even That Is Likely Not Enough”
DESCRIPTION:Abstract:\nLike most of our technologies today\, AI dramatically increases the world’s carbon footprint\, thereby strengthening the severity of the coming downfall of life on the planet. In this talk\, I propose that recent advances in large dimensional mathematics\, and especially random matrices\, could help AI engage in the future economic growth. This being said\, even those mitigating solutions are only temporary in regards to the imminence of collapse\, which calls for drastically more decisive changes in the whole research and industry world. I will discuss these aspects in a second part and hope to leave ample time for discussion. \nHosted by Pia Bideau \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-romain-couillet-university-grenoble-alps-france-random-matrices-could-steer-the-dangerous-path-taken-by-ai-but-even-that-is-likely-not-enough/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2021/11/INS2I-Couillet-D.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20211209T100000
DTEND;TZID=Europe/Berlin:20211216T110000
DTSTAMP:20260619T135227
CREATED:20211125T115420Z
LAST-MODIFIED:20250604T092800Z
UID:11164-1639044000-1639652400@www.scienceofintelligence.de
SUMMARY:Eric J. Johnson (Columbia University\, US)\, “Can We Improve Choices by Changing How Choices Are Posed?”
DESCRIPTION:Abstract:\nChoice architecture suggests that much of what we decide is influenced by that options are presented. This means that the choice environment can encode intelligence that will help (or can hurt) the decision maker. The talk will start by reviewing some results from choice architecture and describe how the environment can affect choice through the choice of strategy and emphasize the role of memory. I will then turn toward developments in studying choice processes including online process tracing techniques and recent developments in the application of eye-tracking using web-based cameras. Finally\, I will talk about applications to presenting consumers and policy makers with information to support sustainable decisions. \nHosted by Oliver Brock\n \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-eric-j-johnson-columbia-university-us-can-we-improve-choices-by-changing-how-choices-are-posed/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2021/11/21.06.01-055-1.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20211202T100000
DTEND;TZID=Europe/Berlin:20211202T110000
DTSTAMP:20260619T135227
CREATED:20211116T143117Z
LAST-MODIFIED:20250604T092818Z
UID:10999-1638439200-1638442800@www.scienceofintelligence.de
SUMMARY:Kate Storrs (Justus Liebig University\, Giessen)\, “Modelling Mid-Level Vision With Unsupervised Learning”
DESCRIPTION:Abstract:\nModels of vision have come far in the past 10 years. Deep neural networks can recognise objects with near-human accuracy\, and predict brain activity in high-level visual regions. However\, most networks require supervised training using ground-truth labels for millions of images\, whereas brains must somehow learn from sensory experience alone. We have been using unsupervised deep learning\, combined with computer-rendered artificial environments\, as a framework to understand how brains learn rich scene representations without ground-truth information about the world. I will show how an unsupervised deep neural network trained on an artificial environment of surfaces that have different shapes\, materials and lighting\, spontaneously comes to encode those factors in its internal representations. Most strikingly\, the model makes patterns of errors in its perception of material that follow\, on an image-by-image basis\, the patterns of errors made by human observers. Unsupervised deep learning may provide a coherent framework for how many perceptual dimensions form\, in mid-level vision and beyond. \nHosted by Martin Rolfs \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-kate-storr-justus-liebig-university-giessen-modelling-mid-level-vision-with-unsupervised-learning/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/png:https://www.scienceofintelligence.de/wp-content/uploads/2021/11/Screenshot-2021-11-16-at-15.27.23.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20210923T100000
DTEND;TZID=Europe/Berlin:20210923T110000
DTSTAMP:20260619T135227
CREATED:20210920T083846Z
LAST-MODIFIED:20240813T093849Z
UID:10644-1632391200-1632394800@www.scienceofintelligence.de
SUMMARY:Tina Klüwer (Science of Intelligence)\, AI Director Science & Startups
DESCRIPTION:Through a talk followed by a discussion and Q&A\, AI Director at Science & Startups Tina Klüwer will explore the joint programmes and resources offered by Berlin’s universities to those wishing to successfully start and develop a company\, also explaining what support is available. \nBIO:\nDr. Tina Klüwer is a recognized expert\, manager and technical ambassador for the topic of Artificial Intelligence and its implementation in business.\nCurrently\, she leads the AI project of Science & Startups\, the network of startup services of the Berlin universities and the Charité Universitätsmedizin Berlin.\nBefore starting the project\, she was founder and CEO of parlamind GmbH\, a company for automation in customer service through AI and language processing. After the successful exit of the business\, she led its four sister companies as technical director. Previously she worked as a researcher at the German Research Center for Artificial Intelligence (DFKI)\, the Bonn University and Freie Universität Berlin\, Germany for over ten years. She received her PhD thesis in computational linguistics from Saarland University.\nTina Klüwer is board member of the German KI Bundesverband (Federal Association for AI)\, Chairwoman of the Technological Sovereignty Advisory Council of the Federal Ministry of Education and Research\, as well as part of the coordination group for AI Standardization by Federal Ministry for Economic Affairs and Energy (BMWI) and DIN. She was also an expert member of the German Bundestag’s two-year Enquete Commission on Artificial Intelligence. \n  \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-tina-kluwer-ai-director-science-startups/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/png:https://www.scienceofintelligence.de/wp-content/uploads/2021/09/Tina-Kluwer.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20210715T100000
DTEND;TZID=Europe/Berlin:20210715T110000
DTSTAMP:20260619T135227
CREATED:20210526T110411Z
LAST-MODIFIED:20250604T095521Z
UID:10233-1626343200-1626346800@www.scienceofintelligence.de
SUMMARY:Dimitri Coelho Mollo (SCIoI)\, “Modelling Intelligence: The Good\, the Bad\, and the Plural”
DESCRIPTION:Abstract:  I argue that artificial intelligence research has been both fuelled and hindered by the use of ‘model tasks’\, that is\, tasks the solution of which are taken to be sufficient for\, or at least indicative of intelligence. Before AI proper\, cybernetics explored model tasks involving basic real-time and world-involving action control aimed at the maintenance of homeostasis\, an approach echoed more recently by the embodied AI movement. Logicist AI\, in contrast\, took as model tasks for intelligence the solution of abstract problems\, such as theorem-proving and proficiency in combinatorially complex games\, chess having pride of place. Connectionist AI – including the current deep learning wave – despite privileging model tasks tied to learning from ‘experience’\, shares this focus on abstract\, disembodied behaviours as key to intelligence\, with particular effort being done in language processing\, categorisation\, and combinatorially complex games\, such as Go. Reliance on model tasks has led to considerable progress in solving those specific tasks\, but against expectation they did not lead to theoretical insights about the nature of intelligence in general\, and how to build it. This outcome\, I argue\, is in part due to the failure of recognising the limited scope of model tasks\, as well as the abstractions and idealisations of real-world intelligent behaviour that they embody. All mainstream frameworks in AI research\, in brief\, focus on circumscribed\, idealised models of intelligent behaviour\, those for which the respective approaches tend to generate cumulative progress and satisfactory solutions. Such models\, however\, abstract or idealise away important features of intelligence\, and\, if unchecked\, close off potentially rewarding paths of research. Bringing to the fore the limitations tied to such model task choices\, as well as the abstractions and idealisation involved in each\, I argue\, opens the way for a more integrative and plural approach to AI. \nThe Zoom Link will be sent the day before the lecture. (Contact communication@scioi.de for specific questions)
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-work-in-progress-dimitri-coelho-mollo-scioi-modelling-intelligence-the-good-the-bad-and-the-plural/
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2020/03/Dimitri1-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20210701T100000
DTEND;TZID=Europe/Berlin:20210701T110000
DTSTAMP:20260619T135227
CREATED:20210526T105942Z
LAST-MODIFIED:20250604T095539Z
UID:10226-1625133600-1625137200@www.scienceofintelligence.de
SUMMARY:Rasmus Rothe\, PhD (Merantix)\, “How To Build a (Deep Tech) Startup”
DESCRIPTION:Abstract: Rasmus Rothe is Co-Founder at Merantix\, the Artificial Intelligent Venture Studio. In this talk he will give insight into how a deep tech startup is built via ideation\, incubation and scaling\, and the specifics and challenges of working with technology AI in the process. \nBIO: Rasmus Rothe is the co-founder and CTO of Berlin-based Merantix\, the world’s first venture studio for AI\, co-initioator of the AI Campus Berlin\, the leading AI community hub in Berlin\, and a renowned deep learning researcher. He has published over 15 academic papers with more than 1000 citations on deep learning while attending Oxford\, Princeton\, and ETH Zurich\, where he received his Ph.D and launched a face recognition service with 150m+ users. In 2019\, he was featured on Forbes “30 under 30”. Rasmus is a founding board member of the German Association of AI\, devising and implementing the national AI strategy in close cooperation with the German government. \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-rasmus-rothe-phd-merantix-how-to-build-a-deep-tech-startup/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/png:https://www.scienceofintelligence.de/wp-content/uploads/2021/05/Screenshot-2021-05-26-at-12.57.59.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20210617T100000
DTEND;TZID=Europe/Berlin:20210617T110000
DTSTAMP:20260619T135227
CREATED:20210429T080929Z
LAST-MODIFIED:20240813T093732Z
UID:10115-1623924000-1623927600@www.scienceofintelligence.de
SUMMARY:Jose Hernandez-Orallo (Valencia/Cambridge)\, "The Generality of Natural and Artificial Intelligence: Task Difficulty as the Elephant in the Room"
DESCRIPTION:Abstract: Understanding and recreating intelligence is possibly the biggest scientific challenge of our time. Evolution has produced organisms that are highly specialised for some cognitive tasks\, whereas others present what has been called general intelligence\, with humans identified as the paragon. Artificial intelligence (AI)\, despite decades of efforts to achieve generality\, is still specialised. It is a major research question to disentangle the notion of general intelligence\, by clearly determining what generality is and how it can be measured for individuals rather than populations. Under limited resources\, we must overhaul the classical yet misleading interpretation of general intelligence as ‘success in all sorts of situations’ and introduce a new view of generality as ‘comprehensive performance up to a level of difficulty’. The degree of generality then refers to the way an agent’s capability is distributed as a function of task difficulty\, according to environmental and cognitive pressures. This dissects the notion of general intelligence into two non-populational measures\, generality and capability. We interpret and apply these measures with humans\, non-human animals and AI systems. The choice of the difficulty function now plays a prominent role in this new conception of generality\, which brings a quantitative tool for shedding light on long-standing questions about the evolution of general intelligence and the evaluation of progress in Artificial General Intelligence. \nHosted by Dimitri Coelho Mollo \nThe Zoom Link will be sent the day before the lecture. (Contact communication@scioi.de for specific questions) \n 
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-jose-hernandez-orallo-valencia-cambridge-the-generality-of-natural-and-artificial-intelligence-task-difficulty-as-the-elephant-in-the-room/
LOCATION:On Zoom
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2021/04/J.H.Orallo-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20210211T100000
DTEND;TZID=Europe/Berlin:20210211T110000
DTSTAMP:20260619T135227
CREATED:20210126T093136Z
LAST-MODIFIED:20250604T095716Z
UID:9602-1613037600-1613041200@www.scienceofintelligence.de
SUMMARY:Alice Auersperg\, “COCKATOOLS: Innovative Tool Use and Manufacture in the Goffin’s Cockatoo”
DESCRIPTION:Finding flexible tool use and manufacture in non-specialized animals\, may contribute to our understanding of the origins of tool-related cognition. Goffin’s cockatoos are Indonesian parrots that originate from a small archipelago in the Moluccas. They are highly opportunist generalists that forage on a large number of different and often patchily distributed or seasonal resources. Accordingly\, they show flexibility and innovativeness during physical problem solving and extractive foraging tasks. Yet more unexpectedly\, in captivity and more recently also in the field we discovered highly flexible tool using and manufacturing abilities rivalling those of the great apes.\nNevertheless\, Goffin’s cockatoos are not dependent on tool obtained resources and lack two ecological predispositions (nest building and food caching) that have been proposed to promote the onset of tool use in birds.\nSo far\, our findings suggest that tool use in this species is associated to opportunism\, extreme extractive foraging and a strong psychological motivation to establish complex object combinations. \nThe Zoom Link will be sent the day before the lecture. (Contact communication@scioi.de for specific questions)
URL:https://www.scienceofintelligence.de/event/thursday-morning-lecture-with-alice-auersperg/
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2019/10/auersperg.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20210204T100000
DTEND;TZID=Europe/Berlin:20210204T230000
DTSTAMP:20260619T135227
CREATED:20210125T164206Z
LAST-MODIFIED:20250604T095741Z
UID:9553-1612432800-1612479600@www.scienceofintelligence.de
SUMMARY:Christa Thöne-Reinecke\, “Ethical Justification of Animal Experiments in Germany”
DESCRIPTION:All animal ethical positions are largely in agreement that animals – as beings capable of suffering – must be morally considered for their own sake and that certain consequences for one’s own actions must be derived from this.\nThis insight has been incorporated into animal protection legislation based on the EU Directive 2010/63.\nGerman legislation requires a reasonable justification of the pain\, suffering\, and harm inflicted on animals.\nFor this reason\, every scientist must demonstrate ethical justifiability of the intended experiment in accordance with the principle of proportionality within the framework of the approval procedure of animal experiments.\nMore specifically\, it must be demonstrated that no alternative method in reaching the project´s aims exists. Furthermore\, the project´s indispensability must be scientifically explained and it must be assigned to a permissible purpose. Study planning must be carried out by implementing statistical methods to reduce the number of animals and their burden to the indispensable level.\nAnimal keeping and medical care must be ensured by the permission to keep and breed animals in the context of a culture of care.\nUltimately\, the expected gain in knowledge must be set in relation to the burden inflicted on the animals and must be ethically justifiable or may even be considered an ethical imperative.\nThe scientist´s proposal and declarations are then revised by the animal welfare officer and\, if applicable\, by the ethics committee of respective institution.\nIt is then further examined by the local authorities and the §15 Commission\, in which ethics experts and animal welfare organizations are actively involved.\nAfter this revision process\, also involving the responsible scientist\, the final examination and approval is carried out by the local authorities.\nIt must be considered that ethical concepts and attitudes of society may be subject to change in the course of time. Hence\, a high degree of transparency is necessary in order to maintain public approval. \nThe Zoom Link will be sent the day before the lecture.
URL:https://www.scienceofintelligence.de/event/thursday-morning-lecture-christa-thone-reinecke/
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2018/11/thoene-reineke_800.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20201203T100000
DTEND;TZID=Europe/Berlin:20201203T110000
DTSTAMP:20260619T135227
CREATED:20201130T132146Z
LAST-MODIFIED:20240813T105558Z
UID:9231-1606989600-1606993200@www.scienceofintelligence.de
SUMMARY:Michael Pauen
DESCRIPTION:BIO: Michael Pauen is a philosopher with a focus on the philosophy of mind. As the academic director of an interdisciplinary graduate school\, he has extensive experience in interdisciplinary research and training. Having a specific interest in philosophical and psychological aspects of human sociality\, he will focus on social intelligence both in humans and in artificial systems. \nThe Zoom Link will be sent the day before the lecture. (Contact communication@scioi.de for specific questions)
URL:https://www.scienceofintelligence.de/event/thursday-mornng-lecture-michael-pauen/
LOCATION:On ZOOM (Contact communication@scioi.de for link)
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2018/11/pauen_800.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20201112T100000
DTEND;TZID=Europe/Berlin:20201112T113000
DTSTAMP:20260619T135227
CREATED:20201102T113930Z
LAST-MODIFIED:20240813T105714Z
UID:9084-1605175200-1605180600@www.scienceofintelligence.de
SUMMARY:Heiko Hamann\, Minimize Surprise in Robots: An Innate Motivation for Collective Behavior
DESCRIPTION:Minimize Surprise in Robots: An Innate Motivation for Collective Behavior \nAfter a quick overview of other related research projects in my lab (bio-hybrid systems\, swarm performance\, collective decision-making)\, I will present our work on minimize surprise for multi-robot systems. Each robot has two artificial neural networks\, a world model (“prediction machine”) and a behavioral module (“action selection network”)\, that are trained concurrently. There is no predefined task\, instead the swarm is rewarded for making correct predictions about future sensory input. As an effect\, robots discover behaviors introducing predictable spatiotemporal sensor patterns. I will present simulated results for flocking\, aggregation\, self-assembly\, construction\, and first results using real-world mobile robots. \nThe Zoom Link will be sent the day before the lecture. (Contact communication@scioi.de for specific questions)
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-heiko-hamann-minimize-surprise-in-robots-an-innate-motivation-for-collective-behavior/
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2019/10/Hamann_800.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20201105T100000
DTEND;TZID=Europe/Berlin:20201105T110000
DTSTAMP:20260619T135227
CREATED:20201102T111116Z
LAST-MODIFIED:20250604T095834Z
UID:9081-1604570400-1604574000@www.scienceofintelligence.de
SUMMARY:Robert Lange (SCIoI): “Learning Not To Learn\, Nature Versus Nurture In Silico”
DESCRIPTION:Abstract: Animals are equipped with a rich innate repertoire of sensory\, behavioral and motor skills\, which allows them to interact with the world immediately after birth. At the same time\, many behaviors are highly adaptive and can be tailored to specific environments by means of learning and exploration. In this work\, we use mathematical analysis and the framework of meta-learning (or ‘learning to learn’) to answer when it is beneficial to learn such an adaptive strategy and when to hard-code a heuristic behavior. We find that the interplay of ecological uncertainty\, task complexity and the agents’ lifetime has crucial effects on the meta-learned amortized Bayesian inference performed by an agent. There exist two regimes: One in which meta- learning yields a learning algorithm that implements task-dependent exploration and a second regime in which meta-learning imprints a purely exploitative and ‘hard-coded’ behavior. Further analysis reveals that non-adaptive behaviors are not only optimal for aspects of the environment that are stable across individuals\, but also in situations where an adaptation to the environment would in fact be highly beneficial\, but could not be done quickly enough to be exploited within the remaining lifetime. Hard-coded behaviors should hence not only be those that always work\, but also those that are too complex to be learned within a reasonable time frame.\nLink: https://arxiv.org/abs/2010.04466 \nThe Zoom Link will be sent the day before the lecture. (Contact communication@scioi.de for specific questions)
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-robert-lange-title-learning-not-to-learn-nature-versus-nurture-in-silico/
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2020/03/Robert1-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20200618T100000
DTEND;TZID=Europe/Berlin:20200618T110000
DTSTAMP:20260619T135227
CREATED:20200528T092941Z
LAST-MODIFIED:20240417T125547Z
UID:7990-1592474400-1592478000@www.scienceofintelligence.de
SUMMARY:Manuel Lopes (hosted by Marc Toussaint): Optimal Behavior Without Optimal Rewards : Artificial Vs Natural
DESCRIPTION:Abstract:\nResearch in robotics and A.I. aims at optimizing very specific task rewards. Intelligent animals have a high degree of curiosity\, and recent\nresults have shown that instrumental reward optimization is a poor explanation for their behavior. We can show that to explain empirical\nresults from animals\, we need to have the drive to optimize reward\, a drive to reduce uncertainty\, and a drive for positive cues. We then show examples in robotics where a more complex reward system provides benefits in learning.\n\nReferences:\nDaddaoua\, N.\, Lopes\, . & Gottlieb\, J. Intrinsically motivated oculomotor exploration guided by uncertainty reduction and conditioned\nreinforcement in non-human primates. Sci Rep 6\, 20202 (2016). https://doi.org/10.1038/srep20202\nLopes\, M.\, Lang\, T.\, Toussaint\, M.\, & Oudeyer\, P. Y. (2012). Exploration in model-based reinforcement learning by empirically estimating learning progress. In Advances in neural information processing systems (pp. 206-214).\n\n***Want to know more about this lecture? Contact us at communication@scioi.de\n\n(Photo by Franck V. on Unsplash)
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-manuel-lopes-hosted-by-marc-toussaint/
LOCATION:On ZOOM (Contact communication@scioi.de for link)
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/jpeg:https://www.scienceofintelligence.de/wp-content/uploads/2020/05/franck-v-zbLW0FG8XU8-unsplash-scaled-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20200528T100000
DTEND;TZID=Europe/Berlin:20200528T110000
DTSTAMP:20260619T135227
CREATED:20200513T133418Z
LAST-MODIFIED:20240813T105459Z
UID:7930-1590660000-1590663600@www.scienceofintelligence.de
SUMMARY:Alan Akbik (SCIoI): Automatically Understanding Human Language: Challenges and Applications
DESCRIPTION:With research in machine learning (ML) and natural language processing (NLP)\, we aim to give machines the ability to understand and use human language. In this talk\, I give a high level introduction of some of the challenges of the field and give an overview of basic NLP tasks (and show some demos). I also introduce the Flair framework – developed by my group together with the open source community – that allows you to use state-of-the-art NLP methods in your research or applications. Time permitting\, I’ll also briefly cover research aspects of the framework\, such as learning word and sentence representations with neural language modeling\, and discuss future directions.
URL:https://www.scienceofintelligence.de/event/thursday-morning-talk-with-alan-akbik-scioi-automatically-understanding-human-language-challenges-and-applications/
LOCATION:On ZOOM (Contact us for Link)
CATEGORIES:Thursday Morning Talk
ATTACH;FMTTYPE=image/png:https://www.scienceofintelligence.de/wp-content/uploads/2020/05/Screen-Shot-2020-05-13-at-15.32.37.png
END:VEVENT
END:VCALENDAR